Refining the causal loop diagram: A tutorial for maximizing the contribution of domain expertise in computational system dynamics modeling.

Abstract

Complexity science and systems thinking are increasingly recognized as relevant paradigms for studying systems where biology, psychology, and socioenvironmental factors interact. The application of systems thinking, however, often stops at developing a conceptual model that visualizes the mapping of causal links within a system, e.g., a causal loop diagram (CLD). While this is an important contribution in itself, it is imperative to subsequently formulate a computable version of a CLD in order to interpret the dynamics of the modeled system and simulate “what if” scenarios. We propose to realize this by deriving knowledge from experts’ mental models in biopsychosocial domains. This article first describes the steps required for capturing expert knowledge in a CLD such that it may result in a computational system dynamics model (SDM). For this purpose, we introduce several annotations to the CLD that facilitate this intended conversion. This annotated CLD (aCLD) includes sources of evidence, intermediary variables, functional forms of causal links, and the distinction between uncertain and known-to-be-absent causal links. We propose an algorithm for developing an aCLD that includes these annotations. We then describe how to formulate an SDM based on the aCLD. The described steps for this conversion help identify, quantify, and potentially reduce sources of uncertainty and obtain confidence in the results of the SDM’s simulations. We utilize a running example that illustrates each step of this conversion process. The systematic approach described in this article facilitates and advances the application of computational science methods to biopsychosocial systems.

Simulating the multicausality of Alzheimer's disease with system dynamics

“…Here, we describe two tests for behavioral validation: a behavior pattern test that compared the model simulations to unseen test data and a structure‐oriented behavior test that compares the simulated interventions to two sets of validation statements. Given that the SDM's key objective involves simulating interventional what‐if scenarios, 20 the structure‐oriented behavior test has precedence.…”

“…Next, we simulated interventions on 15 modifiable risk factors by perturbing them with one standard deviation in 1000 randomly drawn samples from the posterior. We then ranked the risk factors by their effect on cognitive functioning (Alzheimer's Disease Assessment Scale [ADAS‐cog‐13 scale]) after 5 years and compared this ranking to two sets of validation statements 20,24 that were derived from meta‐analyses. The first set contained 44 observational validation statements based on relative risk (RR) ratios from observational data 12 .…”

“…Uncovering and subsequently integrating the web of multicausality that drives AD into a single computational model is equally urgent as it is challenging. 20 Namely, comprehensive computational models are needed to connect current theories and existing data and to test and validate putative causal links. Such models can also assess the relative importance of different risk factors because, unlike non-mechanistic models that can only predict outcomes (e.g., for risk assessment), implementing causal mechanisms allows for simulating interventions.…”

“…As the starting point for developing a computational model, 20 we previously developed a conceptual multicausal mapping in the form of a causal loop diagram (CLD), which summarized expert‐solicited causal links between many of the most important risk factors and pathophysiological mechanisms in sporadic AD 4 . This CLD resulted from the consensus of 15 domain experts through group model building 23 and was supported by a literature review.…”

A cultural-ecosocial systems view for psychiatry

“…These relationships can be mapped by causal loop diagrams (CLD) that aim to capture the links between observable processes ( 240 ). These maps can be used to develop formal quantitative models to reveal dynamics and test the potential impact of interventions, including changes in the configuration of systems—e.g., by altering individual biology or cognition, family interactions, health care systems or other social contingencies ( 241 , 242 ).…”